Role Description We are seeking a hands-on AI Engineer to support the ongoing operation, modernization, and enhancement of AI-enabled applications for the Client. This role will involve working with confidential and regulated financial and examination data. The AI Engineer will primarily work with Python, modern web frameworks such as FastAPI, and cloud services within Microsoft Azure. The candidate will be responsible for developing, maintaining, deploying, and supporting AI-powered applications and backend services. This is an applied technical role in an AI-assisted software development environment. The candidate will work closely with business users, analysts, technical teams, and the Connecticut AI Lab, which provides architectural guidance, shared infrastructure, and engineering support for AI-enabled tools across agencies. Key Responsibilities - Maintain, enhance, and extend existing AI-enabled applications and supporting services. - Develop application logic, backend services, and web APIs using Python and modern web frameworks such as FastAPI. - Build, test, and support applications powered by large language models, including OpenAI GPT, Anthropic Claude, Google Gemini, or similar platforms. - Compare and evaluate AI models based on task performance, accuracy, cost, latency, and reliability. - Configure, deploy, monitor, and maintain applications and services in cloud environments. - Deploy and support services using Microsoft Azure technologies such as Azure App Service, Azure Storage, Azure Key Vault, and related services. - Participate in requirements gathering, design discussions, and project planning sessions with business and technical stakeholders. - Prepare source code, debug issues, correct errors, and maintain software quality. - Use Git-based source control, branching, pull requests, and code review workflows. - Document technical procedures, system workflows, application processes, and deployment steps. - Apply secure development practices when working with confidential, regulated, or sensitive data. - Collaborate with client teams and the clients AI Lab to support AI application development and modernization efforts. - Participate in the evaluation of new AI tools, frameworks, models, and related technologies. Qualifications - Minimum 3 years of Python development experience in a web application environment, including application logic and web APIs. - Minimum 2 years of experience building applications powered by large language models such as OpenAI GPT, Anthropic Claude, Google Gemini, or similar LLM platforms. - Experience building LLM-powered applications that use retrieval-augmented generation, document chunking, embedding generation, semantic or hybrid search, and vector stores. - Minimum 2 years of professional experience improving prompts and evaluating prompt outputs to improve response quality, reduce hallucinations, validate model outputs, and compare models based on accuracy, latency, cost, and reliability, including A/B testing or evaluating model performance for specific tasks. - Experience working with modern web frameworks such as FastAPI, Flask, or Django. - Experience deploying applications and supporting services in cloud environments, including hands-on experience with Microsoft Azure services such as Azure App Service, Azure Storage, Azure Key Vault, Azure Functions, or related services. - Experience using Git-based source control, branching, pull-request-based team workflow, code reviews, and CI/CD practices. - Strong understanding of secure development practices and proper handling of confidential or regulated data. - Experience collaborating directly with both technical and non-technical stakeholders, including business users, analysts, developers, cloud or platform teams, security or compliance teams, and leadership or program stakeholders. - Ability to take a user request from intake through design, development, testing, deployment, and production support with appropriate guidance. - Experience discussing responsible AI, data security, model evaluation, hallucination risk, accuracy, cost, latency, and production reliability with non-technical stakeholders. - Experience using AI-assisted software development tools to write, review, debug, and improve code. - Located within the United States. - Available during Eastern Time business hours. - Available for occasional after-hours support for production issues, deployments, urgent troubleshooting, or system maintenance needs. Preferred Skills - Experience with GitHub or Azure DevOps, especially Azure DevOps. - Experience with AI monitoring, observability, and governance, including tracking prompt performance, model behavior, response quality, drift, logging, and auditability. - Experience presenting technical designs, model comparisons, implementation recommendations, or production support updates to business or agency stakeholders. - Experience with prompt engineering, RAG, embeddings, vector databases, model evaluation, or AI application monitoring.

• Design, build, and productionize foundations for AI applications • Collaborate closely with engineering, product, design, data, operations, security, and business stakeholders • Create secure model access, compliant data pathways, observability, and reusable architecture • Work on AI-assisted sales follow-up, operational defect analysis, and closed-loop feedback systems

Role Description We’re looking for a Senior Software Engineer, Fintech who will help advance the quality, efficiency, and scalability of our engineering team by: - Building product features - Improving developer workflows - Strengthening security controls - Designing shared technical foundations - Helping engineers move faster with better tools and practices Qualifications - 7+ years of professional software engineering experience - 3+ years of experience with Node.js, TypeScript, and MongoDB, or equivalent experience with comparable technologies - Experience contributing directly to compliance activities in a SOC II, ISO-27001, or similar environment - Understanding of backend architecture, API design, distributed systems, and production-grade software development - Experience with AI-assisted development tools such as Cursor, CodeRabbit, Claude, GitHub Copilot, or similar Requirements - Entrepreneurial self-starter with a focus on delivering measurable results - Strong organizational and efficiency skills - Master communicator with the ability to translate complex technical concepts - Understanding of regulatory, security, privacy, and operational expectations in banking and fintech - Tech-forward with analytical thinking skills - Commitment to being a team player Benefits - Comprehensive, people-first benefits package - 100% employer-paid health coverage for employees - 401(k) with 6% employer match and no vesting period - Generous paid time off: 2–4 weeks of vacation based on officer level - Company-paid life insurance - Employee assistance program (EAP) with free counseling, legal, and financial services

• Own agents end to end. • Take a feature from prototype to production: orchestrator and sub-agent design, the tools the agent calls, system prompts, memory, and the response contract the frontend renders from. You write the code that ships. • Own the agent runtime. Design the production runtime to stay fast on the member-facing path and easy to debug when something breaks. • Make the tools trustworthy. Build the tool layer agents depend on, like search grounded in our real catalog, retrieval and reranking, and cart and account actions. • Own safety. Build the layered safety model: input and output guardrails, intent and clarification handling, refusals, and PII boundaries. • Make quality measurable. Push our evaluation work forward: structured checks plus LLM-as-judge, with a review queue for the cases the judges disagree on. • Instrument it. Make agents debuggable in production with per-session and per-turn timelines, tool and guardrail traces, and token and cost visibility. • Turn it into a platform. Take the patterns that work and make them reusable, so the next agent and the next engineer inherit the runtime conventions, the eval scaffolding, and the guardrail defaults instead of starting over. • Make the team better. Set technical direction across the agent codebases and infra, and keep design and code review sharp.